Ana Marusic is a devoted contributor to WAME (and co-editor of the Journal of Global Health). She recently posted a link to a paper in which she was a co-author that deals with the completeness and changes in registered data regarding RCTs. Although there are few RCTs in injury prevention (too few) the findings are important and troubling. It was hoped that the policy of requiring trials to be registered would minimize reporting bias because there was evidence that the results of some (presumably negative) trials were being suppressed. Her paper, published in PLOS online, is based on 149 reports of 152 randomized controlled trials. 9 items from the WHO 20-item Minimum Data Set relevant for assessing trial quality were compared using published vs registry data. The 152 RCTs had a total of 224 missing registry fields and one quarter of the RCTs had major changes in data entries. The “number of RCTs with differences between registered and published data ranged from 21 (13.8%) for Study type to 118 (77.6%) for Target sample size.” They concluded “… the registration data were often not adequate, underwent substantial changes in the registry over time. and differed in registered and published data. The main conclusion was that “Editors need to establish quality control procedures … to contribute to the increased transparency of clinical trials.”
Comment: As I said, there are far too few RCTs in injury prevention but these findings are nevertheless troubling if you accept that there was, indeed, good evidence of reporting bias (I think there was, but largely in relation to pharmaceutical shenanigans) and if you accept that Trial Registration was a sensible solution. Of that I am uncertain and curious to learn what others think. Researchers will not be surprised that there were large differences between what was intended (registered) and what ultimately happened (published). That is the nature of the beast.
The paper can be found at – http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0025258